Memory in humans and deep language models: Linking hypotheses for model augmentation

Omri Raccah, Phoebe Chen, Theodore L. Willke, David Poeppel, Vy A. Vo

Introduction

Modern Transformer models face a significant bottleneck: the quadratic computational cost of self-attention relative to sequence length. This "limited horizon" prevents models from effectively processing long-range dependencies, such as those found in books or long conversations. Memory-Augmented Transformers (MATs) address this by offloading information to external databases. However, determining an optimal "encoding policy"—deciding which specific information to store—remains an open challenge. This research proposes that AI can benefit from "linking hypotheses" derived from cognitive neuroscience, specifically using surprisal as a signal for memory encoding.

The Surprisal Hypothesis

The authors investigate whether surprisal—the negative log-likelihood of an event—serves as a cross-domain signal for memory. In humans, prediction errors (high surprisal) often trigger the brain to encode information into long-term memory. To test this in AI, the researchers compared GPT-2's surprisal scores against human performance in a "cloze task" (word prediction) based on a 7.5-minute spoken story ("Pieman").

Figure 2: Human performance improvement (memory effect) and cosine similarity shifts after story exposure.

Results and Key Insights

The study revealed a robust link between model-based surprisal and human memory. GPT-2 surprisal scores significantly predicted which words participants would remember, showing a strong inverse correlation with cloze performance.

Figure 5: Correlations between GPT-2 surprisal and human memory performance (a) and the resulting memory effect (b).

Crucially, the research debunked a common AI assumption: that "attention weights" are a proxy for importance. The study found that attention weights from GPT-2 showed almost no correlation with human memory performance. This suggests that while attention is vital for immediate processing, it does not necessarily indicate what is worth storing for the long term.

Implications for AI Architecture

These findings provide empirical support for using surprisal as a biologically-inspired encoding policy for MATs. If a model finds a word or concept difficult to predict, that high-surprisal moment acts as a "mental bookmark," signaling that the information should be saved to external memory. This approach moves beyond ad-hoc engineering toward a formal taxonomy of memory-augmented AI that mirrors human psychological processes of encoding and retrieval.

Limitations and Future Work

While promising, the study is limited by the relatively short length of the narrative used (approx. 1,000 tokens). Future research should explore whether the surprisal-memory link persists over much longer durations and more complex tasks, such as free-recall summarization, to further validate these linking hypotheses.

Summary updated

Abstract

The computational complexity of the self-attention mechanism in Transformer models significantly limits their ability to generalize over long temporal durations. Memory-augmentation, or the explicit storing of past information in external memory for subsequent predictions, has become a constructive avenue for mitigating this limitation. We argue that memory-augmented Transformers can benefit substantially from considering insights from the memory literature in humans. We detail an approach for integrating evidence from the human memory system through the specification of cross-domain linking hypotheses. We then provide an empirical demonstration to evaluate the use of surprisal as a linking hypothesis, and further identify the limitations of this approach to inform future research.
BibTeX
@misc{raccah2022memory,
  author = {Omri Raccah and Phoebe Chen and Theodore L. Willke and David Poeppel and Vy A. Vo},
  title = {{Memory in humans and deep language models: Linking hypotheses for model augmentation}},
  year = {2022},
  eprint = {2210.01869},
  archivePrefix = {arXiv},
  url = {https://arxiv.org/abs/2210.01869},
}
Figure